Evolutionary System Identification Using Closed-Loop Data

Abstract

In this paper, a new system identification method is proposed using genetic algorithm(GA), i which the operating data, that is, process closed-loop data, is utilized. Not only system parameters but also control parameters are estimated using the GA. Then, since control parameters are known, the reliability of estimates is evaluated in comparison of the estimates with true control parameters. In order to evaluate the effectiveness of the proposed scheme, it is employed for some chemical process.